Skip to main content

Image and video datasets and models for mxnet deep learning

Project description

# MXbox: Simple, efficient and flexible vision toolbox for mxnet framework.

MXbox is a toolbox aiming to provide a general and simple interface for vision tasks. This project is greatly inspired by [PyTorch]( and [torchvision]( Detailed copyright files are on the way. Improvements and suggestions are welcome.

## Installation
pip install mxbox

## Features
1. Define **preprocess** as a flow

transform = transforms.Compose([
transforms.RandomHorizontalFlip(),, = [ 0.485, 0.456, 0.406 ],
std = [ 0.229, 0.224, 0.225 ]),

PS: By default, mxbox uses `PIL` to read and transform images. But it also supports other backends like `accimage` and `skimage`.

More examples can be found in XXX.

2) Build **DataLoader** in several lines

feedin_shapes = {
'batch_size': 8,
'data': ['data', shape=(8, 3, 32, 32), layout='NCHW')],
'label': ['softmax_label', shape=(8, 1), layout='N')]

dst = Dataset(root='../../data', transform=img_transform, label_transform=label_transform)
loader = DataLoader(dst, feedin_shapes, threads=8, shuffle=True)

Also, common datasets such as `cifar10`, `cifar100`, `SVHN`, `MNIST` are out-of-the-box. You can simply load them from `mxbox.datasets`.

3) Load popular model with pretrained weights

vgg = mxbox.models.vgg(num_classes=10, pretrained=True)
resnet = mxbox.models.resnet152(num_classes=10, pretrained=True)

## Documentation

Under construction, coming soon.

## TODO list

1) Efficient multi-thread reading (Prefetch wanted

2) Common Models preparation.

3) More friendly error logging.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
mxbox-0.0.22-py2.py3-none-any.whl (33.2 kB) Copy SHA256 hash SHA256 Wheel py2.py3

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN SignalFx SignalFx Supporter DigiCert DigiCert EV certificate StatusPage StatusPage Status page